This is a proposal to study the effects of chronic cocaine use and withdrawal on athe amplitude of the P%) mid-latency auditory evoked response (MLAER). Evidence from the literature suggest that both the addictive influence of cocaine and the amplitude of the p%) MLAER are dopamine mediated. Our long-term objective is to define the utility of the P50 measurement in identifying patients who are at increased risk for relapse. By defining the effects of chronic use and of withdrawal of cocaine on the P50 amplitudes we can then plan to study the relapse predicting value of this phenomenon. Our specific hypotheses for the current study are; a) chronic cocaine use adversely affects the amplitude of the P50 component of the MLAERs; and b) the magnitude of changes in the P50 wave correlates with the degree of severity of cocaine dependence. To test the above two hypotheses, we plan to study 60 cocaine addicted subjects. Patents who are being admitted to a 5 week cocaine dependency treatment will be recruited. After careful evaluation of inclusion and exclusion criteria, subjects will have their P50 responses recorded weekly throughout their hospital stay. Clinical and cocaine craving scales will be completed for correlation with P50 changes. Additionally 60 normal age, sex, and education matched controls will be studies. Through multivariate analysis the effects of use of cocaine, degree of use of cocaine or the duration of abstinence, on the amplitude of the P50 component will be studies. We postulate that a decreased P50 amplitude could be reflecting a state of vulnerability of the central nervous system to being flooded by external stimuli. The data gathered in this study will be used to plan subsequent studies of the relapse predicting value of the cocaine induced p50 changes.